Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    avoidance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     transforming
    0.63
     disrupted
    0.58
     tightly
    0.58
     describe
    0.56
     Ding
    0.56
     shaping
    0.56
    沿着
    0.56
    ক্ট
    0.56
    对
    0.55
     mij
    0.54
    POSITIVE LOGITS
    avoid
    0.97
    Avoid
    0.89
     Avoid
    0.88
     éviter
    0.87
     avoid
    0.84
    Avoiding
    0.82
     evitare
    0.82
     avoidance
    0.81
     menghindari
    0.75
     evitar
    0.74
    Act Density 0.109%

    No Known Activations